Download
text classification and na ve bayes n.
Skip this Video
Loading SlideShow in 5 Seconds..
Text Classification and Na ï ve Bayes PowerPoint Presentation
Download Presentation
Text Classification and Na ï ve Bayes

Text Classification and Na ï ve Bayes

160 Views Download Presentation
Download Presentation

Text Classification and Na ï ve Bayes

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

  1. Text Classification and Naïve Bayes The Task of Text Classification

  2. Is this spam?

  3. Who wrote which Federalist papers? 1787-8: anonymous essays try to convince New York to ratify U.S Constitution: Jay, Madison, Hamilton. Authorship of 12 of the letters in dispute 1963: solved by Mosteller and Wallace using Bayesian methods James Madison Alexander Hamilton

  4. Male or female author? • By 1925 present-day Vietnam was divided into three parts under French colonial rule. The southern region embracing Saigon and the Mekong delta was the colony of Cochin-China; the central area with its imperial capital at Hue was the protectorate of Annam… • Clara never failed to be astonished by the extraordinary felicity of her own name. She found it hard to trust herself to the mercy of fate, which had managed over the years to convert her greatest shame into one of her greatest assets… S. Argamon, M. Koppel, J. Fine, A. R. Shimoni, 2003. “Gender, Genre, and Writing Style in Formal Written Texts,” Text, volume 23, number 3, pp. 321–346

  5. Positive or negative movie review? • unbelievably disappointing • Full of zany characters and richly applied satire, and some great plot twists • this is the greatest screwball comedy ever filmed • It was pathetic. The worst part about it was the boxing scenes.

  6. What is the subject of this article? MeSH Subject Category Hierarchy • Antogonists and Inhibitors • Blood Supply • Chemistry • Drug Therapy • Embryology • Epidemiology • … MEDLINE Article ?

  7. Text Classification Assigning subject categories, topics, or genres Spam detection Authorship identification Age/gender identification Language Identification Sentiment analysis …

  8. Text Classification: definition • Input: • a document d • a fixed set of classes C ={c1, c2,…, cJ} • Output: a predicted class c C

  9. Classification Methods: Hand-coded rules • Rules based on combinations of words or other features • spam: black-list-address OR (“dollars” AND“have been selected”) • Accuracy can be high • If rules carefully refined by expert • But building and maintaining these rules is expensive

  10. Classification Methods:Supervised Machine Learning • Input: • a document d • a fixed set of classes C ={c1, c2,…, cJ} • A training set of mhand-labeled documents (d1,c1),....,(dm,cm) • Output: • a learned classifier γ:d c

  11. Classification Methods:Supervised Machine Learning • Any kind of classifier • Naïve Bayes • Logistic regression • Support-vector machines • k-Nearest Neighbors • …

  12. Text Classification and Naïve Bayes The Task of Text Classification

  13. Text Classification and Naïve Bayes Naïve Bayes (I)

  14. Naïve Bayes Intuition • Simple (“naïve”) classification method based on Bayes rule • Relies on very simple representation of document • Bag of words

  15. The bag of words representation I love this movie! It's sweet, but with satirical humor. The dialogue is great and the adventure scenes are fun… It manages to be whimsical and romantic while laughing at the conventions of the fairy tale genre. I would recommend it to just about anyone. I've seen it several times, and I'm always happy to see it again whenever I have a friend who hasn't seen it yet. γ( )=c

  16. The bag of words representation I lovethis movie! It's sweet, but with satiricalhumor. The dialogue is greatand the adventure scenes are fun… It manages to be whimsicaland romanticwhile laughingat the conventions of the fairy tale genre. I would recommendit to just about anyone. I've seen it severaltimes, and I'm always happyto see it againwhenever I have a friend who hasn't seen it yet. γ( )=c

  17. The bag of words representation: using a subset of words x lovexxxxxxxxxxxxxxxxsweetxxxxxxxsatiricalxxxxxxxxxxxxxxxxxxxxxgreatxxxxxxxxxxxxxxxxxxxxxxxxxxfunxxxxxxxxxxxxxxxxxwhimsicalxxxxromanticxxxxlaughingxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxrecommendxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxseveralxxxxxxxxxxxxxxxxxxxxxxhappyxxxxxxxxxagainxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx γ( )=c

  18. The bag of words representation γ( )=c

  19. Bag of words for document classification ? Test document Machine Learning NLP Garbage Collection Planning GUI parser language label translation … planning temporal reasoning plan language... learning training algorithm shrinkage network... parser tag training translation language... garbage collection memory optimization region... ...

  20. Text Classification and Naïve Bayes Naïve Bayes (I)

  21. Text Classification and Naïve Bayes Formalizing the Naïve Bayes Classifier

  22. Bayes’ Rule Applied to Documents and Classes • For a document dand a class c

  23. Naïve Bayes Classifier (I) MAP is “maximum a posteriori” = most likely class Bayes Rule Dropping the denominator

  24. Naïve Bayes Classifier (II) Document d represented as features x1..xn

  25. Naïve Bayes Classifier (IV) • O(|X|n•|C|) parameters How often does this class occur? • Could only be estimated if a very, very large number of training examples was available. We can just count the relative frequencies in a corpus

  26. Multinomial Naïve Bayes Independence Assumptions Bag of Words assumption: Assume position doesn’t matter Conditional Independence: Assume the feature probabilities P(xi|cj) are independent given the class c.

  27. Multinomial Naïve Bayes Classifier

  28. Applying Multinomial Naive Bayes Classifiers to Text Classification positions  all word positions in test document

  29. Text Classification and Naïve Bayes Formalizing the Naïve Bayes Classifier

  30. Text Classification and Naïve Bayes Naïve Bayes: Learning

  31. Sec.13.3 Learning the Multinomial Naïve Bayes Model • First attempt: maximum likelihood estimates • simply use the frequencies in the data

  32. Parameter estimation • Create mega-document for topic j by concatenating all docs in this topic • Use frequency of w in mega-document fraction of times word wiappears among all words in documents of topic cj

  33. Sec.13.3 What if we have seen no training documents with the word fantasticand classified in the topic positive (thumbs-up)? Zero probabilities cannot be conditioned away, no matter the other evidence! Problem with Maximum Likelihood

  34. Laplace (add-1) smoothing for Naïve Bayes

  35. Multinomial Naïve Bayes: Learning • From training corpus, extract Vocabulary • Calculate P(wk|cj)terms • Textj single doc containing all docsj • Foreach word wkin Vocabulary nk # of occurrences of wkin Textj • Calculate P(cj)terms • For each cjin C do docsjall docs with class =cj

  36. Text Classification and Naïve Bayes Naïve Bayes: Learning

  37. Text Classification and Naïve Bayes Naïve Bayes: Relationship to Language Modeling

  38. Generative Model for Multinomial Naïve Bayes c=China X1=Shanghai X2=and X3=Shenzhen X4=issue X5=bonds

  39. Naïve Bayes and Language Modeling • Naïvebayes classifiers can use any sort of feature • URL, email address, dictionaries, network features • But if, as in the previous slides • We use only word features • we use all of the words in the text (not a subset) • Then • Naïvebayes has an important similarity to language modeling.

  40. Sec.13.2.1 Each class = a unigram language model • Assigning each word: P(word | c) • Assigning each sentence: P(s|c)=Π P(word|c) Class pos 0.1I 0.1love 0.01 this 0.05fun 0.1film … I love this fun film 0.1 0.1 .05 0.01 0.1 P(s | pos) = 0.0000005

  41. Sec.13.2.1 I love this fun film 0.1 0.1 0.01 0.05 0.1 0.2 0.001 0.01 0.005 0.1 Naïve Bayes as a Language Model • Which class assigns the higher probability to s? Model pos Model neg 0.2 I 0.001love 0.01this 0.005fun 0.1 film 0.1I 0.1love 0.01this 0.05fun 0.1film P(s|pos) > P(s|neg)

  42. Text Classification and Naïve Bayes Naïve Bayes: Relationship to Language Modeling

  43. Text Classification and Naïve Bayes Multinomial Naïve Bayes: A Worked Example

  44. Priors: P(c)= P(j)= 3 Choosing a class: P(c|d5) P(j|d5) 1 4 4 • 3/4 * (3/7)3 * 1/14 * 1/14 ≈ 0.0003 Conditional Probabilities: P(Chinese|c) = P(Tokyo|c) = P(Japan|c) = P(Chinese|j) = P(Tokyo|j) = P(Japan|j) = (5+1) / (8+6) = 6/14 = 3/7 (0+1) / (8+6) = 1/14 • 1/4 * (2/9)3 * 2/9 * 2/9 ≈ 0.0001 (0+1) / (8+6) = 1/14 (1+1) / (3+6) = 2/9 (1+1) / (3+6) = 2/9 (1+1) / (3+6) = 2/9

  45. Naïve Bayes in Spam Filtering • SpamAssassin Features: • Mentions Generic Viagra • Online Pharmacy • Mentions millions of (dollar) ((dollar) NN,NNN,NNN.NN) • Phrase: impress ... girl • From: starts with many numbers • Subject is all capitals • HTML has a low ratio of text to image area • One hundred percent guaranteed • Claims you can be removed from the list • 'Prestigious Non-Accredited Universities' • http://spamassassin.apache.org/tests_3_3_x.html

  46. Summary: Naive Bayes is Not So Naive • Very Fast, low storage requirements • Robust to Irrelevant Features Irrelevant Features cancel each other without affecting results • Very good in domains with many equally important features Decision Trees suffer from fragmentation in such cases – especially if little data • Optimal if the independence assumptions hold: If assumed independence is correct, then it is the Bayes Optimal Classifier for problem • A good dependable baseline for text classification • But we will see other classifiers that give better accuracy

  47. Text Classification and Naïve Bayes Multinomial Naïve Bayes: A Worked Example

  48. Text Classification and Naïve Bayes Precision, Recall, and the F measure

  49. The 2-by-2 contingency table

  50. Precision and recall • Precision: % of selected items that are correctRecall: % of correct items that are selected